Commit Graph

15 Commits

Author SHA1 Message Date
funman300 4d48cad4e3 fix(engine): hide pile markers under cards — kill the gray-corner artifact
Player feedback after the border-drop fix did NOT close the
"gray corners" complaint: "I do not see anything change." The
border was a real artifact, but the *visible* gray came from a
different source.

Root cause: pile markers are 8%-alpha-white sprites sized to
the card area, sitting at `Z_PILE_MARKER = -1.0` beneath every
card. Composited against the dark play surface, the marker's
effective colour is ≈`#272727` — visibly gray. When a card
(rounded corners, opaque body) sits on top, the marker's
rectangular fill bleeds through the 4 small triangular regions
where the card's rounded corner curves cut away from the card's
bounding rectangle. That bleed-through is the "gray L" the
player saw at each card corner.

Fix: hide pile-marker sprites for any pile that has a card on
top. New `sync_pile_marker_visibility` system runs each Update
tick, guarded by `game.is_changed()` so the work skips on idle
frames. Iterates `(&PileMarker, &mut Visibility)` and sets
`Hidden` for occupied piles, `Inherited` for empty.

This implements the *documented* invariant declared in the
module-level doc comment ("Pile markers ... remain visible only
where a pile is empty") that was previously not enforced —
markers always rendered. Strictly speaking this is a
documentation-vs-implementation drift fix, not a behaviour
change.

### Why the border-drop fix didn't address this

The border drop changed the SVG stroke and removed *one* source
of corner artifacts (anti-aliased red/near-white stroke fading
through gray). It correctly drifted 52 face hashes. But the
visible gray at corners came from a *different* layer — the
pile-marker sprite *behind* the card, not the card stroke
itself. Right test target, wrong visible-artifact target.
Two layers, two fixes; this commit closes the second.

### Test

New `pile_markers_hide_when_pile_is_occupied` pins the
post-deal state: 8 markers hidden (stock + 7 tableau), 5
markers visible (waste + 4 foundations). 1192 passing
(+1 from prior 1191).

Workspace clippy clean.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-08 12:49:13 -07:00
funman300 651f4060e6 refactor(engine): migrate table_plugin chrome to Terminal tokens
- Promote `marker_colour` to module-level const PILE_MARKER_DEFAULT_COLOUR
  and re-export it. cursor_plugin::MARKER_DEFAULT now imports the const
  directly, replacing the prior duplicated literal kept in sync only by
  doc comment. Drift becomes a compile error instead of a stale claim.
- Empty-tableau "K" placeholder text now uses TEXT_PRIMARY at 0.35 alpha
  (was raw `Color::srgba(1.0, 1.0, 1.0, 0.35)`) so it picks up the
  Terminal off-white foreground.
- HINT_PILE_HIGHLIGHT_COLOUR retuned from bright `srgb(1.0, 0.85, 0.1)`
  to the design-system STATE_WARNING token (`#ddb26f`). Spelled as a
  literal because Alpha::with_alpha is not yet const on stable; a new
  test pins the RGB to STATE_WARNING so a palette swap can't drift the
  two apart silently.
- The existing "is gold" character test was hardcoded to the old bright
  palette (red ≥ 0.9). Loosened to "warmer than cool" + ranges that the
  Terminal muted gold satisfies, with exact-RGB tracking handled by the
  new STATE_WARNING test.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-07 18:32:03 -07:00
funman300 95df5421c9 feat(core): unlock foundations — Foundation(u8) slots, suit derived from contents
Standard Klondike behaviour: any Ace can land in any empty foundation,
and that slot then claims the suit until the pile empties. The
previous PileType::Foundation(Suit) variant pre-assigned each of the
four foundations to a fixed suit ("C / D / H / S" placeholders) and
rejected mismatched Aces — non-standard and (per the smoke-test
feedback) confusing.

Replaces the variant payload with a slot index Foundation(u8) (0..=3)
and derives the claimed suit from the bottom card via a new
Pile::claimed_suit() method. The bottom card is, by construction,
the Ace that established the claim; using it directly eliminates an
entire class of "stuck claim after undo" bugs that a separate
claimed_suit field would have introduced.

can_place_on_foundation drops its suit parameter — the rule reduces
to "empty pile accepts any Ace; non-empty pile accepts the next
rank up of the bottom card's suit." Iteration sites across
input_plugin, cursor_plugin, selection_plugin, card_plugin,
auto_complete_plugin, game_plugin, layout, and hud_plugin all swap
the four-suit list for `(0..4u8).map(PileType::Foundation)`.

next_auto_complete_move now prefers a slot whose claimed_suit matches
the candidate card before falling back to the first empty slot for
an Ace — so the same suit consistently auto-targets the same slot
across the whole game, matching player expectations.

The HUD selection label and the hint toast read claimed_suit() and
fall back to "Foundation N" / "move to foundation" only when the
slot is empty. Empty foundation pile markers no longer render the
suit-letter children — they're plain translucent rectangles, matching
empty tableau placeholders.

Save-format invalidation: GameState gains a schema_version field
(serde-default to 1 for back-compat parsing of old files), the
constant is bumped to 2, and load_game_state_from rejects mismatched
schemas. Old in-progress saves silently fall through to "fresh game
on launch" — the user accepted this loss given the mechanic change.
Stats / progress / achievements / settings live in separate files,
contain no PileType data, and are unaffected.

9 new tests pin the contract:
- Pile::claimed_suit returns None for empty / non-foundation, Some
  for non-empty foundation
- Any Ace lands in the first empty foundation; successive Aces
  distribute across slots 0..3
- Claim drops when the slot is emptied via undo
- Auto-complete picks the slot with a matching claim, not the first
  empty slot
- A v1-format game_state.json is rejected; sibling stats save/load
  is unaffected

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 22:17:17 +00:00
funman300 4b9d008be2 refactor(workspace): sweep low-risk clippy::pedantic findings
Conservative cleanup pass — applied only the high-signal pedantic
lints whose fixes either remove genuine waste or read more naturally,
skipping anything stylistic that would bloat the diff.

- map_unwrap_or: 29 .map(...).unwrap_or(...) sites collapsed to
  .map_or / .is_some_and / .map_or_else equivalents
- uninlined_format_args: 7 production format!/write!/println! sites
  rewritten to the inline-argument style; assert! sites in test code
  intentionally untouched
- match_same_arms: 2 redundant arms collapsed where the bodies were
  identical and the merger didn't obscure intent

Public API is unchanged. No dependencies added or removed. The
pedantic warning count dropped from 840 to 807 (-33). Out-of-scope
findings — needless_pass_by_value on Bevy Res params, false-positive
explicit_iter_loop on Bevy Query iterators, items_after_statements
inside test mods, and the "ask before changing" merge logic in
solitaire_sync — were intentionally deferred.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-01 02:46:32 +00:00
funman300 366fd6d127 fix(engine): snap cards directly on window resize
CI / Test & Lint (push) Failing after 19s
CI / Release Build (push) Has been skipped
on_window_resized was firing StateChangedEvent on every WindowResized
event. That ran sync_cards_on_change → update_card_entity, which
inserts a CardAnim slide tween for every card whose target moves >1
unit. During a corner drag the resize fires every frame, retargeting
the slide each time from the cards' current mid-tween positions, so
cards never reach steady state — the visible "snap back and forth"
jitter reported during the 2026-04-29 smoke test.

Replace the StateChangedEvent emit with a direct snap path:

- Add LayoutSystem::UpdateOnResize SystemSet in layout.rs so cross-
  plugin ordering is explicit (Bevy's automatic conflict-based order
  only forces non-parallel execution, not a particular order).
- table_plugin::on_window_resized: drop the StateChangedEvent emit;
  mark the system in_set(LayoutSystem::UpdateOnResize). It already
  snaps backgrounds and pile markers directly, so this aligns cards
  with the same instant-snap policy.
- card_plugin: new snap_cards_on_window_resize system listens for
  WindowResized, runs .after(LayoutSystem::UpdateOnResize), writes
  fresh transforms via the existing card_positions() helper, and
  removes any in-flight CardAnim. It also reapplies the stock-empty
  indicator so the "↺" label's font_size (derived from
  layout.card_size.x) still rescales on resize.

Other StateChangedEvent listeners — start_settle_anim,
detect_auto_complete, clear_selection_on_state_change, check_no_moves,
reset_hint_cycle_on_state_change, clear_right_click_highlights — no
longer fire spuriously on resize. They should not fire on a layout
change anyway; that was a pre-existing minor bug masked by the
jitter.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-29 22:44:08 +00:00
funman300 fbe984cf64 feat(engine): switch asset loading to AssetServer with xCards artwork
CI / Test & Lint (push) Failing after 19s
CI / Release Build (push) Has been skipped
Replace compile-time include_bytes!() embedding for card faces, backgrounds,
and font with runtime AssetServer::load() calls. Swap in 52 xCards @2x PNGs
(LGPL-3.0) as card face assets and xCards bicycle_blue as back_0.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-29 20:06:02 +00:00
funman300 7cda2a9f1a fix(engine): resolve all clippy warnings introduced by PNG asset pipeline
CI / Test & Lint (push) Failing after 1m34s
CI / Release Build (push) Has been skipped
- Collapse nested-if patterns into let-chains across 13 plugins (42 instances)
- Add #[allow(clippy::too_many_arguments)] to 5 Bevy systems in card_plugin
  and input_plugin where ECS parameter count exceeds the lint threshold
- Gate Theme import in table_plugin under #[cfg(test)] — only used by
  test-only colour helpers; removing the unconditional import silences the
  unused-import lint without breaking the test suite
- Wrap ButtonInput<MouseButton> in Option<> in update_input_platform so that
  tests using MinimalPlugins (no InputPlugin) no longer panic on startup

All 789 tests pass; cargo clippy --workspace -- -D warnings is clean.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-29 03:35:41 +00:00
funman300 18ac5adef5 feat(engine): art pass — PNG assets, custom font, and keyring v4 upgrade
Art pass (Phase 4):
- Generate placeholder PNG assets: face.png, back_0–4.png, bg_0–4.png via
  solitaire_assetgen gen_art binary (16×16 RGBA, embedded via include_bytes!)
- Add FiraMono-Medium font (assets/fonts/main.ttf) embedded at compile time
- Add FontPlugin: loads font at startup, exposes FontResource; gracefully
  falls back to default handle when Assets<Font> absent (MinimalPlugins tests)
- Wire CardImageSet into card_plugin: face/back PNGs replace solid-colour
  sprites when available; tests continue using colour fallback via MinimalPlugins
- Wire BackgroundImageSet into table_plugin: bg PNGs replace solid-colour
  background; empty set inserted when Assets<Image> absent in tests
- Fix hint highlight system (input_plugin): tint sprite.color directly instead
  of replacing the whole Sprite (which would discard the image handle)
- Export FontPlugin, FontResource, CardImageSet from solitaire_engine::lib
- Register FontPlugin in solitaire_app before other plugins

Dependency upgrades (latest releases):
- keyring "2" → keyring "4" + keyring-core "1" (v4 split architecture into
  separate core library crate)
- auth_tokens.rs: Entry::new now returns Result; delete_password →
  delete_credential; NoDefaultStore error variant handled
- solitaire_app: add keyring::use_native_store(true) at startup for Linux
  Secret Service / macOS Keychain / Windows Credential Store selection

ARCHITECTURE.md: fix Edition 2025→2021, update asset pipeline section,
add FontPlugin/CardImageSet/BackgroundImageSet to plugin and resource tables,
update Section 14 to reflect actual include_bytes!() rendering approach,
add Decision Log entries for embedded PNG and font decisions

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-29 00:30:55 +00:00
funman300 648cd44387 chore(deps): migrate to Bevy 0.17
- Event/EventReader/EventWriter renamed to Message/MessageReader/MessageWriter
- add_event → add_message for all 67 call sites
- ScrollPosition changed to tuple struct ScrollPosition(Vec2)
- CursorIcon import moved from bevy::winit::cursor to bevy::window
- WindowResolution::from((f32,f32)) removed — use (u32,u32) tuple
- World::send_event → World::write_message in test code

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-28 13:04:44 -07:00
funman300 8cd28cfb29 feat(engine): right-click highlight timer and visual hint glow (#5, #6)
Task #5: Add RightClickHighlightTimer(1.5 s) so destination highlights
auto-despawn after 1.5 s. Existing clear-on-state-change and
clear-on-pause logic still fires early when a move is made or the game
is paused. Three unit tests cover timer countdown behaviour.

Task #6: Add HintVisualEvent emitted on H key. Source card gets
HintHighlight + HintHighlightTimer(2 s) for a yellow glow. Destination
PileMarker gets HintPileHighlight with a gold tint (Color::srgb(1.0,
0.85, 0.1)) that restores the original colour when the 2 s timer
expires. Five unit tests cover timer expiry and colour invariants.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-28 17:36:23 +00:00
funman300 c3ee7c45a7 feat(engine): card visual improvements — flip animation, foundation/tableau placeholders, drag shadow
Task #34: CardFlipAnim component + start_flip_anim/tick_flip_anim systems animate revealed
cards by squashing scale.x to 0 then expanding back to 1 (2×0.08 s). Skipped at Instant speed.

Task #35: spawn_pile_markers now adds a Text2d child (S/H/D/C, 45% alpha) on Foundation
markers so the suit is visible while the pile is empty.

Task #43: Tableau pile markers get a "K" Text2d child (35% alpha) indicating only Kings land
on empty columns.

Task #38: update_drag_shadow system maintains a single ShadowEntity while dragging — a
card_w+8 × card_h+8 dark semi-transparent sprite at z−1 behind the top dragged card.

Also fixed pre-existing clippy/compiler errors in hud_plugin, pause_plugin, stats_plugin,
cursor_plugin, and settings_plugin (missing imports, too-many-arguments, doc formatting).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-27 19:03:59 +00:00
funman300 fe23e89971 test(engine): add advance_elapsed saturation and theme colour pure-function tests
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-27 05:17:15 +00:00
root 299e0c6a94 feat(engine): cosmetic selectors applied, stats screen expanded, daily goals enforced
- Card backs: selected_card_back index maps to distinct Color values in card rendering
- Backgrounds: selected_background index applied in TablePlugin alongside theme
- Both re-render immediately on SettingsChangedEvent
- Stats screen now shows Games Lost, Draw 1/3 Wins, and Lifetime Score
- Daily challenge win no longer credited if server-supplied target_score or max_time_secs constraints are not met

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-27 01:46:52 +00:00
root 3c01cef5f3 feat(engine): implement Draw Mode, Theme, and Music Volume settings
Settings panel "coming soon" stubs replaced with live controls:

- Draw Mode toggle (Draw 1 / Draw 3): new games read draw_mode from
  SettingsResource instead of the previous game's mode. Falls back to
  the current game's mode in headless/test contexts where SettingsPlugin
  is absent.

- Theme selector (Green → Blue → Dark → Green): SettingsChangedEvent
  drives TablePlugin's background Sprite colour so the table re-colours
  immediately without a restart.

- Music Volume [−]/[+]: dedicated kira sub-tracks created for SFX and
  music on startup. SFX sounds are routed to the SFX track; the music
  track exists for future ambient audio. Both volumes are set on
  SettingsChangedEvent and at startup.

Also fixed: time_attack timer_expiry test double-fires when
MinimalPlugins time delta is nonzero — removed the intermediate
0.001-remaining update step.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-04-26 23:53:48 +00:00
funman300 d92b4a8648 feat(engine): add layout, LayoutResource, and TablePlugin
compute_layout is a pure function that maps window size to card size and
the 13 pile positions, with clamping at the 800x600 minimum and seven
tableau columns horizontally aligned with stock/waste (cols 0,1) and the
four foundations (cols 3,4,5,6). TablePlugin spawns a 2D camera, a felt
background sprite, and 13 translucent pile-marker sprites, and
repositions them on WindowResized. Plugin registers WindowResized
explicitly so it works under MinimalPlugins in tests.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-04-23 16:18:24 -07:00